write-roxygen-docs
About
This Claude Skill generates comprehensive roxygen2 documentation for R packages, covering functions, datasets, classes, and methods while following tidyverse style. It handles standard tags, cross-references, examples, and NAMESPACE entries automatically. Use it when documenting new exports, internal helpers, or fixing R CMD check notes related to documentation.
Quick Install
Claude Code
Recommendednpx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/write-roxygen-docsCopy and paste this command in Claude Code to install this skill
Documentation
Write Roxygen Docs
Complete roxygen2 docs → R fns, datasets, classes.
Use When
- New exported fn → docs
- Internal helper fns
- Pkg datasets
- S3/S4/R6 classes + methods
- Fix doc-related
R CMD checknotes
In
- Required: R fn|dataset|class to doc
- Optional: Related fns → cross-ref (
@family,@seealso) - Optional: Export fn?
Do
Step 1: Fn Docs
Roxygen comments directly above fn:
#' Compute the weighted mean of a numeric vector
#'
#' Calculates the arithmetic mean of `x` weighted by `w`. Missing values
#' in either `x` or `w` are handled according to the `na.rm` parameter.
#'
#' @param x A numeric vector of values.
#' @param w A numeric vector of weights, same length as `x`.
#' @param na.rm Logical. Should missing values be removed? Default `FALSE`.
#'
#' @return A single numeric value representing the weighted mean.
#'
#' @examples
#' weighted_mean(1:5, rep(1, 5))
#' weighted_mean(c(1, 2, NA, 4), c(1, 1, 1, 1), na.rm = TRUE)
#'
#' @export
#' @family summary functions
#' @seealso [stats::weighted.mean()] for the base R equivalent
weighted_mean <- function(x, w, na.rm = FALSE) {
# implementation
}
Got: Complete roxygen w/ title, desc, @param per param, @return, @examples, @export.
If err: Unsure tag → ?roxygen2::rd_roclet. Common omission @return → CRAN required for all exports.
Step 2: Essential Tags
| Tag | Purpose | Required for export? |
|---|---|---|
#' Title | First line, one sentence | Yes |
#' Description | Paragraph after blank line | Yes |
@param | Parameter documentation | Yes |
@return | Return value description | Yes (CRAN) |
@examples | Usage examples | Strongly recommended |
@export | Add to NAMESPACE | Yes, for public API |
@family | Group related functions | Recommended |
@seealso | Cross-references | Optional |
@keywords internal | Mark as internal | For non-exported docs |
Got: Required tags ID'd. Exports have @param, @return, @examples, @export minimum.
If err: Tag unfamiliar → roxygen2 docs for usage + syntax.
Step 3: Doc Datasets
Create R/data.R:
#' Example dataset of city temperatures
#'
#' A dataset containing daily temperature readings for major cities.
#'
#' @format A data frame with 365 rows and 4 variables:
#' \describe{
#' \item{date}{Date of observation}
#' \item{city}{City name}
#' \item{temp_c}{Temperature in Celsius}
#' \item{humidity}{Relative humidity percentage}
#' }
#' @source \url{https://example.com/data}
"city_temperatures"
Got: R/data.R has roxygen blocks per dataset w/ @format describing structure + @source for provenance.
If err: R CMD check warns undocumented dataset → ensure quoted string ("city_temperatures") exactly matches obj name saved w/ usethis::use_data().
Step 4: Doc Pkg
Create R/packagename-package.R:
#' @keywords internal
"_PACKAGE"
## usethis namespace: start
## usethis namespace: end
NULL
Got: R/packagename-package.R exists w/ @keywords internal + "_PACKAGE" sentinel. devtools::document() generates man/packagename-package.Rd.
If err: R CMD check reports missing pkg doc page → verify file R/<packagename>-package.R + contains "_PACKAGE".
Step 5: Special Cases
Fns w/ dots in names (S3 methods):
#' @export
#' @rdname process
process.myclass <- function(x, ...) {
# S3 method
}
Reuse docs w/ @inheritParams:
#' @inheritParams weighted_mean
#' @param trim Fraction of observations to trim.
trimmed_mean <- function(x, w, na.rm = FALSE, trim = 0.1) {
# implementation
}
No visible binding fix w/ .data pronoun:
#' @importFrom rlang .data
my_function <- function(df) {
dplyr::filter(df, .data$column > 5)
}
Got: Special cases (S3 methods, inherited params, .data pronoun) documented correctly. @rdname groups S3 methods. @inheritParams reuses params w/o duplicate.
If err: R CMD check warns "no visible binding for global variable" → #' @importFrom rlang .data or utils::globalVariables() last resort.
Step 6: Generate Docs
devtools::document()
Got: man/ updated w/ .Rd files per documented obj. NAMESPACE regenerated w/ correct exports + imports.
If err: Roxygen syntax errs. Common: unclosed brackets in \describe{}, missing #' prefix, invalid tag names. Re-run devtools::document() after fix.
Check
- Every exported fn has
@param,@return,@examples -
devtools::document()runs no errs -
devtools::check()no doc warnings -
@familytags group correctly - Examples run no errs (
devtools::run_examples())
Traps
- Missing
@return: CRAN requires all exports doc return value - Examples need internet/auth: Wrap
\dontrun{}w/ comment why - Slow examples:
\donttest{}for examples that work but slow for CRAN - Markdown in roxygen: Enable
Roxygen: list(markdown = TRUE)in DESCRIPTION - Forget
devtools::document(): Man pages generated, not hand-written
→
create-r-package— initial pkg setup including roxygen configwrite-testthat-tests— test fns you docwrite-vignette— long-form docs beyond fn refsubmit-to-cran— doc requirements for CRAN
GitHub Repository
Related Skills
content-collections
MetaThis skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.
polymarket
MetaThis skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.
creating-opencode-plugins
MetaThis skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
